- Horizon AI
- Posts
- OpenAI Disbands Yet Another Safety Team 👀
OpenAI Disbands Yet Another Safety Team 👀
Can a reporter’s AI clone fool his colleagues?
Gina Acosta
October 26, 2024
In partnership with
Welcome to another edition of Horizon AI,
OpenAI has disbanded a team responsible for developing safeguards around advanced AI systems, raising safety concerns. Meanwhile, Meta has just announced a new partnership with Reuters, marking the latest alliance between a major tech company and a news provider.
Let’s jump right in!
Read Time: 3.5’ min
Here's what's new today in the Horizon AI
Chart of the week: Companies Acquiring the Most Startups
OpenAI Is Disbanding Yet Another AI Safety Team
Meta Strikes Its First Big AI News Deal With Reuters
Free Resources
AI tools to check out
Video of the week
TOGETHER WITH WRITER
Escaping AI POC purgatory: Techniques for enterprise AI engineers
From POC to production: practical strategies for bringing AI to production successfully
Expert insights: Led by Sam Julien, Director of Developer Relations
Real-world examples: Drive business impact with AI
Chart of the week
Companies Acquiring the Most Startups
The data covers 2,532 corporate-startup M&A transactions completed between 2000 and the first half of 2024, limited to Fortune Global 500 companies.
North American companies acquire startups four times more often than European firms and 7-10 times more than Asian firms. Alphabet (Google) leads with 222 acquisitions, followed by Microsoft and Cisco.
AI News
OPENAI
OpenAI Is Disbanding Yet Another AI Safety Team
OpenAI has disbanded its AGI Readiness Team, which focused on safety measures for advanced AI systems, raising concerns among industry experts about the company's direction in AI development.
Details:
The team focused on the safety of AGI, defined by OpenAI as AI systems that can autonomously perform various human tasks. Team members have been reassigned to other roles within the company.
Miles Brundage, OpenAI's departing Senior Advisor for AGI Readiness, has expressed serious concerns about this development in a detailed public statement as he announces his exit from the company.
In May, OpenAI disbanded its "superalignment" team, focused on AI's existential risks. Co-founder and former chief scientist Ilya Sutskever and co-founder Jan Leike both resigned that month.
The disbandment of these teams indicates that OpenAI might not view advanced AI as an immediate concern.
META
Meta Strikes Its First Big AI News Deal With Reuters
Facebook's owner, Meta, has signed a multi-year agreement with Reuters to provide real-time answers to user questions about news and current events through its Meta AI chatbot.
Details:
Reuters will be compensated for its content appearing in Meta’s AI chatbot, accessible through Facebook, Instagram, WhatsApp, and Messenger, with links to Reuters stories beginning to appear for US users on Friday.
The deal is Meta's first of its kind and reflects the growing trend of news outlets partnering with AI companies.
It's not clear if the partnership includes a licensing component that would allow Meta to use Reuters' journalism to train its large language model, Llama.
Some tech giants and news organizations are starting to find common ground, especially as AI companies face pressure to prevent their chatbots from spreading misinformation.
Whenever you are ready, here's how we can help you:
Book an Audit call with our AI Automation agency: Our clients saved 40+ hours in their first month. Limited spots available. (Your business must be making a minimum of $5K/month)
Resources
💡 How — and why — this brand created an ad campaign entirely with AI
🔍 What is Character AI? What parents need to know
👀 AI detects mysterious detail hidden in a famous Raphael masterpiece
AI Tools to check out
🔊 Wondercraft: AI-powered audio studio that makes it easy to create audio content, like ads, podcasts and meditations, without recording.
📽 Vidify: Turn product photos into AI videos with image-to-video generation.
📷 Pixmaker: AI-generated professional photos and videos to boost business revenue.
💥 Figflow: Transform Figma designs into user stories in minutes.
💡 Napkin: Turns your text into visuals so sharing your ideas is quick and effective.
Video of the week
Can a reporter’s AI clone fool his colleagues?
A demonstration from Joe Tidy of how deepfake clones are now able to join and communicate in live meetings.
Victims are increasingly reporting generative AI scams, including a Hong Kong incident where an AI clone allegedly deceived staff into a $25m loss. While concerns grow, companies like Zoom see potential for AI clones to attend meetings on our behalf.
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻